Re: Linked Stuff [was Re: RDF's challenge]

On 6/11/13 6:29 PM, Luca Matteis wrote:
>
> On Wed, Jun 12, 2013 at 12:17 AM, Kingsley Idehen 
> <kidehen@openlinksw.com <mailto:kidehen@openlinksw.com>> wrote:
>
>     This conversation is about being flexible about how we promote
>     Linked Data to folks that have long tuned out the letters R-D-F.
>
>
> I agree we need to make it more compelling to users with as little 
> jargon as possible. And I also agree that many times RDF is 
> misinterpreted to be some form of XML and developers really hate XML.
>
> But we can't ignore the fact that RDF is clearly defined in all of the 
> definitions I found so far of "Linked Data" as the pillar standard 
> data making this distributed Web of Data function.
>
> So of course I agree we can explain it to people without mentioning 
> RDF [1], but RDF is still at the heart of Linked Data's definitions 
> the same as HTTP is at the heart of the definition of the "Web".
>
> 1. http://www.genebanks.org/tutorials/the-power-of-linked-data

Again, RDF combined with Linked Data principles is a wonderful thing. At 
the same time, we don't need to fall into the "mantra recital" trap that 
ultimately undermine the virtues of RDF by overloading it. Blogic [1] is 
where the combination of RDF comes into play.

RDF doesn't lie at the core of Linked Data since you can craft web-like 
structured data (what Linked Data actually is) that conforms to all 
versions of TimBL's memes.


For instance, assuming you accept that RDF didn't invent the triple 
(3-tuple), what RDF knowledge to I need to save the following to a 
document that's published to a Web accessible location:

<> <#type> <#Document> .
<> <#mentions> <#i> .
<#i> <#type> <#Person> .
<#> <#name> "Kingsley Idehen.
<#i> <#email> <mailto:kidehen@openlinksw.com> .

Ah! You could claim "that's Turtle notation and its associated with RDF" 
. I could push back and claim that its just CSV or plain text. I can 
express the above knowing that I am in the act of structured data 
representation without any interest in RDF or knowledge of it.

RDF's distinguishing characteristics kick-in if I want to apply 
reasoning to the data above by leveraging relations semantics explicitly 
described in a vocabulary. Remember, RDF is 100% self-describing, you 
all of the basic semantics of RDF are actually expressible in RDF.

Now if you want to take the latest is the greatest view and claim that 
Linked Data is based on RDF, due to its inclusion in the revised Linked 
Data meme. Then why isn't it also based on SPARQL? Remember, point #3 of 
the meme gives equal standing to RDF and SPARQL. Thus, why isn't Linked 
Data not based on RDF plus SPARQL?



Links:

1. http://www.slideshare.net/PatHayes/blogic-iswc-2009-invited-talk -- 
blogic .

-- 

Regards,

Kingsley Idehen	
Founder & CEO
OpenLink Software
Company Web: http://www.openlinksw.com
Personal Weblog: http://www.openlinksw.com/blog/~kidehen
Twitter/Identi.ca handle: @kidehen
Google+ Profile: https://plus.google.com/112399767740508618350/about
LinkedIn Profile: http://www.linkedin.com/in/kidehen